# Neon RiddleNode Intent-Matrix Surrogate Report

## Plain-English Summary

The Neon RiddleNode loop can now run several small mix-change requests and add a conservative development-review surrogate on top of the objective receipts.

The useful catch is workflow speed. When the receipts are clean, Codex can mark a conservative candidate as approved for development review so iteration can continue without waiting for immediate human listening. That surrogate approval is explicitly weaker than taste approval.

## What Changed

- LilArcade PR #534 added `npm run proof:sequencer:riddlenode-intent-matrix-surrogate`.
- The command runs the RiddleNode intent matrix against `https://riddlenode.com/neon-lab/`.
- The matrix still runs shared gates once, then checks bass, guitar, and chord "down a little" packets.
- After objective receipts pass, a mixing-canon surrogate reviews each packet for conservative development approval.
- The surrogate does not apply candidates automatically and does not claim any candidate sounds better.

## Verification

Live RiddleNode surrogate matrix:

- status: `intent_matrix_ready_for_review`
- shared gates: `local_gate_ready`
- bass request: `bass -0.05`, supported `3`, rejected `0`
- guitar request: `guitar -0.02`, supported `3`, rejected `0`
- chord request: `chord -0.035`, supported `3`, rejected `0`
- deterministic findings: `0`
- review warnings: `0`
- surrogate status: `approved_for_development_review`
- surrogate approvals: `3`
- needs human review count: `0`
- recommended development candidate: `guitar -0.02`

Local checks:

- `node --test scripts/__tests__/neonRatchetIntentMatrix.test.mjs scripts/__tests__/neonRatchetBatch.test.mjs`
- `npm run test:neon`
- `npm run proof:sequencer:riddlenode-intent-matrix-surrogate`

GitHub checks:

- LilArcade PR #534 CI passed at head `834fe34`.

## Proof Boundary

This proves that the Neon lab can batch bounded public-target claim-candidate loops, generate human-review packets, and add a conservative development-review surrogate from objective receipts.

It also shows the low-level subtle candidate cap working on the live target: the quiet chord lane now keeps the three-iteration "down a little" path inside supported candidates rather than spending an iteration on an over-large rejected cut.

It does not prove subjective mix quality, listener preference, the best musical candidate, or that any candidate should be promoted automatically.